sign in sign up
<<
Home , Bacterial cellulose, Diguanylate cyclase, Gluconacetobacter, Gluconokinase, Phosphofructokinase 2

www.nature.com/scientificreports

OPEN Complete genome analysis of Gluconacetobacter xylinus CGMCC 2955 for elucidating bacterial Received: 8 November 2017 Accepted: 3 April 2018 biosynthesis and Published: xx xx xxxx metabolic regulation Miao Liu1, Lingpu Liu1, Shiru Jia1, Siqi Li1, Yang Zou2 & Cheng Zhong1

Complete genome sequence of Gluconacetobacter xylinus CGMCC 2955 for fne control of bacterial cellulose (BC) synthesis is presented here. The genome, at 3,563,314 bp, was found to contain 3,193 predicted genes without gaps. There are four BC synthase operons (bcs), among which only bcsI is structurally complete, comprising bcsA, bcsB, bcsC, and bcsD. Genes encoding key in glycolytic, pentose , and BC biosynthetic pathways and in the tricarboxylic acid cycle were identifed. G. xylinus CGMCC 2955 has a complete glycolytic pathway because sequence data analysis revealed that this strain possesses a (pf)-encoding gene, which is absent in most BC-producing strains. Furthermore, combined with our previous results, the data on of various sources (monosaccharide, , and acetate) and their regulatory mechanism of action on BC production were explained. Regulation of BC synthase (Bcs) is another efective method for precise control of BC biosynthesis, and cyclic diguanylate (c-di-GMP) is the key activator of BcsA–BcsB subunit of Bcs. The quorum sensing (QS) system was found to positively regulate , which decomposed c-di-GMP. Thus, in this study, we demonstrated the presence of QS in G. xylinus CGMCC 2955 and proposed a possible regulatory mechanism of QS action on BC production.

Bacterial cellulose (BC), naturally produced by several species of , is a strong and ultra-pure form of cellulose1. It has the same chemical structure as plant cellulose but is devoid of and lignin2. Until now, many researchers have endeavoured to control BC production to realise its potential applications. Most applications have been implemented by optimizing the carbon source and culture conditions3,4. Additionally, functionalisation or modifcation of BC has mainly been achieved via chemical or mechanical modifcations of the cellulose matrix5,6. Nowadays, synthetic biology enables model microorganisms to be easily reprogrammed with modular DNA code to perform a variety of new tasks for useful purposes7. Tis approach may allow for more fnely tuned control over BC synthesis by gene regulation. Nonetheless, only a few attempts at genetic engineering have been made for BC-producing , and the toolkit for genetic modifcation has hardly been described8. Tus, it is essential for researchers to gain genomic insights into various BC-producing strains. So far, several BC-producing Gluconacetobacter () species have been sequenced, and the genomes of K. xylinus E259, K. medellinensis NBRC 3288 (ref.10), K. nataicola RZS0111, and Gluconobacter oxy- dans DSM 3504 were sequenced completely. Gluconacetobacter xylinus (formerly Acetobacter xylinum) is one of the most commonly studied species because of its ability to produce relatively large amounts of BC from a wide range of carbon and sources in liquid culture3,4,12 and has become the most popular strain so far for manufacturing of various materials, including paper13, food packaging14, medical materials15, and cell culture16. It is difcult to obtain high productivity by means of G. xylinus in a large-scale fermentation system owing to its low yield under agitated culture. Nonetheless, BC secreted by G. xylinus has unique properties including high

1Key Laboratory of Industrial Fermentation , Ministry of Education, Tianjin University of Science and Technology, Tianjin, 300457, P. R. China. 2Tianjin Jialihe Livestock Group Co., Ltd, JinWei Road, Beichen District, Tianjin, 300402, P. R. China. Correspondence and requests for materials should be addressed to C.Z. (email: chzhong. [email protected])

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 1 www.nature.com/scientificreports/

Genome size (bp) 3,563,314 Gap N (bp) 0 GC content (%) 63.29 Total genes 3,139 Non-coding RNA 117 rRNA 45 tRNA 57 others 15 Repeat DNAs SINEs 13 LINEs 9 LTR 0 DNA elements 6 Small RNA 41 Satellites 0 Simple repeats 409 Low complexity 38

Table 1. General features of G. xylinus CGMCC 2955 genome. SINEs: short interspersed elements. LINEs: long interspersed nuclear elements. LTR: long terminal repeat.

quality in terms of mechanical strength, water-holding capacity, crystallinity, biodegradability, and biocompat- ibility17. G. xylinus CGMCC 2955 originates from the fermentation substrates of vinegar18. It has been studied as a model organism for BC production for decades and has been successfully applied to production of wound dressings19,20. In the current study, we present a complete genome sequence of G. xylinus CGMCC 2955 to provide background information for genetic engineering so that precise control of BC biosynthesis can be achieved on the basis of the metabolic pathway that we proposed previously3. A comparison of the arrangement and composition of BC synthase operons (bcs) with those of other BC-producing strains was performed. Furthermore, the relation between the bcs operon and quorum sensing (QS) system – a widely studied system for population regulation – is discussed. Results Sequencing showed that the genome is 3,563,314 bp long with GC content of 63.29%. One scafold was obtained without gaps (Table 1). Gene prediction and annotation of the G. xylinus CGMCC 2955 genome resulted in 3,193 predicted genes. Amongst 117 non-coding RNAs, there are 45 rRNAs, 57 tRNAs, and 15 other RNAs. In the repeat DNAs, simple repeats turned out to be the most abundant repeat families, accounting for 0.51% of the G. xylinus CGMCC 2955 genome, followed by small RNA, accounting for 0.20% of the genome. Non–long terminal repeats were also identifed in the genome and contain 13 short interspersed nuclear elements and nine long interspersed nuclear elements. Besides, six DNA transposons (fve DNA/hAT-Ac specimens and one DNA/ TcMar-Tigger) were detected. No long terminal repeats or satellites were observed. Te cluster of orthologous groups (COG) function-based classifcation of G. xylinus CGMCC 2955 genes revealed 23 functional groups, in which 19.96% of the genes are functionally uncharacterised (Fig. 1). Te second and third largest functional groups were ‘transcription’ and ‘ transport and metabolism’; they consisted of 988 and 985 genes, respectively.

Carbon source metabolism. A total of 50 genes of transporters, 13 genes of symporters, and 50 genes of permeases were identifed in the genome of G. xylinus CGMCC 2955. Tey are responsible for the transport of sugars (e.g. , arabinose, and galactose), sugar acids (e.g. gluconate, α-ketoglutarate, and galactonate), amino acids, spermidines, vitamin B12, phosphate, metal ions, and lipopolysaccharide. G. xylinus CGMCC 2955 is capable of producing BC from several carbon sources3. Generally, glucose (Glc) is metabolised through the Embden–Meyerhof–Parnas (EMP) pathway, which has proven to be incomplete in most BC-producing stains because of the lack of a gene encoding phosphofructokinase (PFK; EC 2.7.1.11)11,21,22. Of note, according to the gene sequence analysis, the G. xylinus CGMCC 2955 genome possesses a gene encoding phosphofructokinase B (pfB, CT154_09360), while there is no gene encoding phosphofructokinase A (pfA). Te other two key enzymes are (GK) and pyruvate (PK), whose encoding gene locus tags are CT154_11525 and CT154_15810, respectively. Te pentose phosphate pathway (PPP) is another efective glucose metabolic pathway in G. xylinus. In the G. xylinus CGMCC 2955 genome, genes encoding glucose-6-phosphate dehydrogenase (6pgd, CT154_14710), glucose-6-phosphate 1-dehydrogenase (6pgd-1, CT154_14205 and CT154_05345), 6-phosphogluconolactonase (6pgl, CT154_14190), 6-phosphogluconate dehydrogenase (6pgad, CT154_14210), ribose-5-phosphate A (rpi A, CT154_14185), ribulose-phosphate 3-epimerase (rpe, CT154_13290), transketolase (tkt, CT154_07695, CT154_08455, CT154_12910, and CT154_14220), and transal- dolase (tsd, CT154_07690 and CT154_14215) were identifed. Gluconic acid (GlcA) is an important by- of BC production in G. xylinus3,23. It was found to be produced from glucose by glucose dehydrogenase (gd, CT154_01250 and CT154_06665) and can be converted to 6-phosphogluconate (GlcA6P) by (GntK). Tere are two genes encoding GntK (CT154_14180 and CT154_03305), and the latter was hypothesised

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 2 www.nature.com/scientificreports/

Figure 1. Cluster of orthologous groups (COG) classifcation of functions.

Figure 2. BC production from ethanol and acetate as the sole or supplementary carbon sources. (None: no carbon source, Act: acetate, Eth: ethanol, Glu: glucose).

to encode a thermosensitive GntK. Te tricarboxylic acid cycle (TCA) has three key enzymes – citrate synthase (CT154_06205), isocitrate dehydrogenase (CT154_11710 and CT154_00570), and 2-oxoglutarate dehydrogenase (CT154_14680) – and the genes coding for these enzymes are all present in G. xylinus CGMCC 2955. (FRU) and (GLY) can considerably contribute to a BC yield just as glucose can3. Via by fruc- tokinase (frk, CT154_06355), fructose was found to be converted to fructose-6-phosphate (F6P). When glyc- erol was utilised for BC production, it was frstly converted to glycerol 3-phosphate (GLY3P) by (glyk, CT154_10145 and CT154_10550), and then was transformed into phosphate (DHAP) and glyceraldehyde-3-phosphate (GAP) by glycerol-3-phosphate dehydrogenase (G3PD) and triosephos- phate isomerase (TPI), entering the EMP pathway and PPP. Teir encoding gene locus tags were found to be CT154_10150 and CT154_05740, respectively. When ethanol (ETH) or acetate (AC) served as a sole carbon source for BC production, as shown in Fig. 2, both contributed to a BC production of 0.4 g/l. Glucose was then added as the main carbon source, and either ethanol or acetate were used as supplementary carbon sources. As compared to glucose as the sole carbon source, the ethanol and acetate supplementation increased the BC yield to 1.26-fold and 1.31-fold, respectively. When ethanol is added as the supplementary carbon source, Acetobacter xylinum BPR 3001 A can synthesise BC with a higher yield24. In BPR 3001 A, ethanol is frst converted to acetaldehyde by dehydrogenase (ADH), and then acetaldehyde is converted to acetate under the action of aldehyde dehydrogenase (ALDH). Afer that, acetyl-coenzyme A (ACCoA) is produced from acetate. In the genome of G. xylinus CGMCC 2955, the gene

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 3 www.nature.com/scientificreports/

sequences of Adh (CT154_02385, CT154_08135, CT154_09950, CT154_11925, and CT154_13530) and Aldh (CT154_00245, CT154_02130, CT154_06670, and CT154_12865) were found to be available. In , acetate can be utilised as a sole carbon source in the phosphotransacetylase- pathway (Pta-Ack), in which phosphate acetyltransferase (PATs) and acetate kinase (ACK) are required25, encoded by CT154_12395 and CT154_15860 in G. xylinus CGMCC 2955.

Energy metabolism. Te BC yield enhancement by ethanol and acetate supplementation has been reported to be related to (ATP) production24. Tere are two patterns for ATP production: level phosphate production through glucose metabolism and oxidative phosphate production via a reduced form of nicotinamide-adenine dinucleotide (NADH) (or reduced favin adenine dinucleotide, FADH2) generated by glucose metabolism and transferred to electron acceptors. Glucose metabolism is the main energy source of G. xylinus CGMCC 2955. Under aerobic conditions, is the electron acceptor from NADH and FADH2 through the electron transport chain. Under these conditions, energy production is mainly based on oxidative , sup- plemented by substrate level phosphorylation26. Under anaerobic culture conditions, nitrate acts as the electron acceptor in Enterobacter sp. FY-07, which is able to produce BC with a high yield under anaerobic conditions. It is reduced to nitrite by accepting the electrons transferred through the electron transport chain, and nitrite is then reduced to ammonia27. Nitrate reductase and nitrite reductase are involved in this process, and in the pres- ent study, only a nitrite reductase gene (CT154_11275) was identifed in the G. xylinus CGMCC 2955 genome. Moreover, genes encoding ADH (CT154_02385, CT154_05910, CT154_08135, CT154_09950, CT154_11925, and CT154_13530) and (ldh, CT154_01275 and CT154_10640) were identifed in the G. xylinus CGMCC 2955 genome. Tey are responsible for the production of NADH when oxygen supply is defcient. Tis fnding suggested that G. xylinus CGMCC 2955 has the potential to produce sufcient energy in hypoxic environments.

BC synthesis. In G. xylinus, glycolytic intermediate glucose-6-phosphate is the origin of substrate synthe- sis for BC production. It is isomerised to glucose-1-phosphate (G1P) by phosphoglucomutase (PGM; encoded by CT154_06830), which then reacts with UTP, forming uridine-5′-phosphate-α-D-glucose (UDPG) under the action of UDP-glucose pyrophosphorylase (UGPase; encoded by CT154_06835)28. Finally, cellulose is synthe- sised by a BC synthase (Bcs) complex containing subunits BcsA, BcsB, BcsC, and BcsD, which are encoded by three (bcsAB, bcsC, and bcsD) or four (bcsA, bcsB, bcsC, and bcsD) genes11,29. Te Bcs complex spans the outer and inner cell membranes to synthesise and extrude chains, which are then assembled into fbrils that are further assembled into a ribbon30. BcsA is the catalytic subunit of the Bcs complex28,31. BcsB is an auxiliary subunit that interacts with BcsA via its C-terminal transmembrane helix and regulates BC synthesis by interacting with cyclic diguanylate (c-di-GMP)32. Te BcsA–BcsB complex is sufcient for BC synthesis and translocation31. BcsC is believed to play important roles in cell membrane pore formation for BC secretion32. BcsD may control the crystallisation of cellulose into nanofbrils5. Tere can be one or several bcs operons in a single organism, and the sequence identity of genes encoding the same subunits may not be that high. Moreover, the composition and arrangement of bcs operons are diverse in the same or two diferent organisms. Additionally, among the bcs operons in the same genome, only one or some of the operons perform key functions in BC synthesis33. In the G. xylinus CGMCC 2955 genome, a total of four bcs operons were identifed. As shown in Fig. 3, their compositions were diferent among the operons. Operon bcs I was the only structurally complete one, con- taining bcsA (CT154_12695), bcsB (CT154_12690), bcsC (CT154_12685), and bcsD (CT154_12680). In addi- tion to the above genes, cmcax (CT154_12705) and ccpax (CT154_12700) were identifed upstream, and the bglxA (CT154_12675) gene was located downstream in operon bcs I. Gene cmcax encodes endo-β-1,4-glucanase (CMCax). When antibodies to recombinant CMCax are added to the culture medium, the formation of cellulose fbre is severely inhibited34. Te CcpAx protein (also known as ORF-2), encoded by ccpax, functions as a mediator of protein–protein interactions and is important for localisation of the Bcs complex to the cell membrane. Te bglxa gene encodes a β-glucosidase. CMCax and β-glucosidase both can hydrolyse tangled glucan chains when there is a failure in chain arrangement and are both crucial for BC synthesis35. Operons bcs II and bcs III consist of bcsA, bcsB, and bcsC, where BcsA and BcsB are encoded by one gene. Operon bcs II (bcsA, bcsB, and bcsC) was found to be encoded by CT154_04230 and CT154_04245, with bcsX (CT154_04235) and bcsY (CT154_04240) in the middle of bcsB and bcsC. Te products of bcsX and bcsY have not yet been characterised although Bcs Y was predicted to function as a transacylase, participating in the production of acetyl cellulose or similarly mod- ifed polysaccharides36. Operon bcs III (bcsA, bcsB, and bcsC) turned out to be encoded by CT154_05660 and CT154_05665. Operon bcs IV uniquely contains only bcsA and bcsB (CT154_03760). Sequence alignment and analysis of bcsA and bcsB in the four operons indicated that the identity was very low between them (Table 2). Operon bcs I was found to be structurally the same as the sole bcs operon in G. xylinus ATCC 53582. Te sequence alignment of each gene is presented in Fig. 3b. It suggests that bcsA in bcs I (bcsAI) genes in G. xylinus CGMCC 2955 and G. xylinus ATCC 53582 share 67.68% and 66.97% nucleotide sequence identity and amino acid sequence identity, respectively. For bcsBI, the respective identity values were 67.48% and 64.02%. For bcsCI, the respective identity values were 66.49% and 63.83%. For bcsDI, the respective identity levels were 77.49% and 76.92%. Tere is only one bcs operon in G. xylinus ATCC 53582. Te high identity of operon bcs I sequences between G. xylinus CGMCC 2955 and G. xylinus ATCC 53582, and its complete structure suggested that bcs I may be deeply involved in BC synthesis in G. xylinus CGMCC 2955. Tese results provide a founda- tion for genetic manipulation for research regarding which bcs operon is essential for BC synthesis in G. xylinus CGMCC 2955.

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 4 www.nature.com/scientificreports/

Figure 3. An overview of the G. xylinus CGMCC 2955 genome. (a) Te circle represents (from outside in): circle 1, genome; circle 2, G + C content; circle 3, genes on forward (red) and reverse (green) strands; circle 4, distribution of genes encoding non-coding RNAs; circle 5, long terminal repeats in the genome. (b) Te arrangement of bcs operons and sequence alignment of the bcs I operon and G. xylinus ATCC 53582 bcs operon.

bcs I A-B bcs II A-B bcs III A-B bcs IV A-B bcs I A-B 100 57.31 53.36 56.79 bcs II A-B 57.31 100 56.99 67.87 bcs III A-B 53.36 56.99 100 55.87 bcs IV A-B 56.79 67.87 55.87 100

Table 2. sequence alignment of bcs A-B in four bcs operons of G. xylinus CGMCC 2955.

Regulation of BC synthesis by c-di-GMP and by the QS system. Te BcsA–BcsB complex in Bcs is activated by c-di-GMP37. Te latter is synthesised from (GTP) by (DGC) and is degraded by two (PDEs), thereby regenerating 5′- (5′-GMP; that can be used for GTP synthesis) and forming a biosynthetic and degradative pathway: the c-di-GMP cycle2,38. In the G. xylinus CGMCC 2955 genome, there is only one dgc gene (CT154_08975), while there are several pde genes, including CT154_00885, CT154_01735 – CT154_01745, CT154_08940, CT154_08945, CT154_08975, and CT154_14645. Besides, CT154_05705 and CT154_14650 are hypothesised to encode DGC/PDE.

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 5 www.nature.com/scientificreports/

Figure 4. Te reported vertical streaking method with bacterial biosensors. (a) G. xylinus CGMCC 2955. (b) P. aeruginosa PAK (positive control).

Tese genes together control intracellular c-di-GMP levels. In Gluconacetobacter intermedius, pde expression was shown to be positively regulated by the GinI–GinR QS system39,40. Te presence of a QS system can be determined by detection of chemical signals called autoinducers41. Tese signalling are produced from common metabolites such as fatty acids, anthranilate, and S-adenosylmethionine42–44. In the current study, the presence of signalling molecules was identifed by a vertical streaking well difusion method. tumefaciens A136 allows for detection of a broad range of acyl-homoserine lactones (AHLs) with acyl chain lengths from 4 to 12 as well as 3-unsubstituted AHLs45. As shown in Fig. 4, the A. tumefaciens A136 biosensor turned blue in (a) and (b) indicating that G. xylinus CGMCC 2955 and aeruginosa PAK both synthesise a QS AHL autoinducer signalling (Fig. 4a,b)46. Tis fnding revealed that G. xylinus CGMCC 2955 is capable of producing AHLs, and the latter are usually produced by the LuxI–LuxR QS system in gram-negative bacteria47. Genes luxI and luxR are normally homologs of ginI and ginR. Nevertheless, only the luxR (CT154_10285) gene was identifed in G. xylinus CGMCC 2955. Discussion To control the production of BC for diferent purposes, researchers manipulate the cell culture process. Te car- bon source was reported to be one of the most efective factors that contribute to diferences in BC production and structural characteristics12,48,49. Zhong et al. revealed that G. xylinus CGMCC 2955 can produce BC from various carbon sources, including glucose, fructose, and glycerol3. Distinct BC network structures were obtained from these carbon sources, resulting from diferent metabolic processes, including BC production and cell growth rates. Te metabolic network necessary to study the metabolism of diferent carbon sources has already been built3. In the present study, genome sequence analysis revealed the metabolic characteristics of G. xylinus CGMCC 2955. Most BC-producing strains do not possess the EMP pathway because they lack PFK activity50. Nevertheless, genome sequence analysis indicated that G. xylinus CGMCC 2955 has pfB (CT154_09360), which encodes PFK II51. PFK I and II both convert fructose-6-phosphate (F6P) to fructose-1,6-diphosphate (F16P). In E. coli, ~90% of the activity is attributed to PFK I, a well-known allosteric . Te remaining activity is PFK II52. Terefore, in contrast to other BC-producing microorganisms, G. xylinus CGMCC 2955 has a complete EMP pathway. Moreover, ~33% of glucose enters the PPP and 17% entered the TCA cycle in G. xylinus CGMCC 29553. In the genome of G. xylinus CGMCC 2955, all genes encoding enzymes of the PPP were found, confrming that glucose could be metabolised by the PPP. Besides, fux analysis suggested that 92.85% of glucose is fuxed into gluconic acid3. Liu et al. speculated that gluconic acid can serve as a second carbon source when glucose is exhausted in G. xylinus CGMCC 295523. Te presence of gluconate kinase–encoding gene gntK (CT154_14180, CT154_03305) – coupled with a fux analysis that indicated that more than a half of gluconic acid (52.83%) was converted to glucose-6-phosphate – suggested that gluconic acid may be utilised as a carbon source for cell growth and BC production through the PPP. Tis fnding is consistent with our previous report showing that dry cell weight and the BC yield increased 2.55- and 14.17-fold, respectively, when gluconic acid was provided as the sole carbon source as opposed to no carbon source53. When ethanol or acetate were used as the sole carbon source, neither of them could contribute to a con- siderable BC yield (Fig. 2). On the other hand, the genes encoding the enzymes of the Pta-Ack pathway are all present in G. xylinus CGMCC 2955. When G. xylinus CGMCC 2955 was incubated in a glucose-containing culture medium, both ethanol and acetate contributed to the increased BC yield. Tese results support other reports24. We believe that ethanol or acetate cannot be utilised as a substrate for BC production owing to the absence of a precursor for BC synthesis. Nevertheless, ethanol or acetate functioned as an energy source for ATP generation through the TCA cycle. Terefore, it was inferred that the promotion of BC production by supple- mentary ethanol or acetate can be attributed to ATP generation, which accelerated the BC synthesis pathway by inhibiting glucose-6-phosphate dehydrogenase activity24. Furthermore, the relation between ATP production and BC synthesis was recently studied in Enterobacter sp. FY-07, which is capable of synthesizing BC under both aerobic and anaerobic culture conditions27. Te BC biosynthesis process in Enterobacter sp. FY-07 was found to be the same as that of G. xylinus. It manifested a high yield of BC even under anaerobic conditions, suggesting

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 6 www.nature.com/scientificreports/

that oxygen was not directly involved in the BC synthesis reaction. Meanwhile, ATP was found to be necessary to activate glucose to UDPG, which is also essential for c-di-GMP biosynthesis. Our recent work indicates that G. xylinus CGMCC 2955 can produce more BC during hypoxia than under atmospheric and oxygen-enriched culturing conditions53. Further research on the efect that oxygen and energy generation have on BC production is a possible next project. The Bcs complex is the key enzyme in BC synthesis. Four bcs operons were identified in the G. xylinus CGMCC 2955 genome. Up to three bcs operons in other BC-producing species have been reported (G. xylinus ATCC 23769 and G. hansenii ATCC 53582)10,54,55, where a high cellulose synthase copy number generally indi- cates contributions to high BC productivity8. Our phylogenetic analysis indicates that bcs II and bcs IV were the most closely related (Table 2) and possibly arose from duplication and subsequent translocation. With this information, we can study the function of diferent bcs operons, e.g. investigate whether operon bcs I, the only structurally complete bcs operon, functions alone. If the genes encoding Bcs subunits were deleted in bcs I, will other bcs operons act as candidate genes? c-di-GMP, an activator of the BcsA–BcsB complex, is synthesised from GTP by DGC and is degraded by PDE2. In G. intermedius, the expression of pde was found to be positively regulated by the GinI–GinR QS system39,40. QS is a microbial cell-to-cell communication process that allows a group of bacterial cells to regulate their gene expression in unison, which is important for implementation of group behaviours such as bioluminescence56, virulence57, phenotypic heterogeneity58, genetic exchange59, and bacterial pathogenicity60. QS systems have been identifed in both gram-negative and gram-positive bacterial species61. Te LuxI–LuxR-type system is common among gram-negative QS bacteria, in which LuxR serves as both the cytoplasmic signalling molecule’s receptor and the transcriptional activator of the lux operon’s QS system, whereas LuxI works as the signalling molecule’s synthase47,62,63. In this study, Fig. 4 illustrates the presence of AHLs in G. xylinus CGMCC 2955. AHLs were syn- thesised by the LuxR–LuxI system. When the AHL signal reaches a threshold concentration, it is bound by LuxR, and this complex activates the transcription of a specifc operon. Nonetheless, in the G. xylinus CGMCC 2955 genome, only luxR (CT154_10285) was identifed. Te absence of luxI suggests that the signalling molecule may be produced by other pathways instead of luxI or that the identity of the luxI sequence in G. xylinus CGMCC 2955 with that in other microorganisms is too low for identifcation of luxI. Te regulation of the QS system by PDE involves the 89 aa protein GinA (target protein) in G. intermedius, where the production of GinA is induced by the QS system. A putative c-di-GMP phosphodiesterase, pde was shown to be GinA inducible and is involved in the repression of oxidative fermentation by the GinI–GinR QS system40. Terefore, because luxI and luxR are homologs of ginI and ginR, and c-di-GMP is an activator of the BcsA–BcsB subunit, we speculated that BC biosynthesis may be regulated by QS by controlling c-di-GMP levels. Bacteria generally depend on QS to regulate cellular processes that are essential for survival and adaptation to changing environments. Although it is still unclear why microorganisms produce BC in nature, BC has been shown to confer access to sufcient oxygen at the liquid–gas interface and high resistance to UV irradiation64. Terefore, there is a possibility that QS participates in the BC synthesis. Various methods have been applied to manipulate the biosynthesis of BC, but little attention has been paid to the regulation of QS during BC produc- tion. Nonetheless, it has been successfully applied to the production of bioflms65. Te BC-regulatory mechanism involved in the relations among QS, c-di-GMP, and BC is worthy of further study. Te regulatory mechanism of action on BC production presented in this study is summarised in Fig. 5. Materials and Methods Cell culture and chromosomal DNA extraction. G. xylinus CGMCC 2955 was isolated from 16 solid fermentation substrates of vinegar by Tianjin University of Science and Technology18. Afer cultivation on a solid medium (25 g/l glucose [0.83 (mol carbon)/l], 10 g/l peptone, 7.5 g/l yeast extract, and 10 g/l Na2HPO4) at 30 °C for 3 days, cells in the medium were washed with normal saline and centrifuged at 4000 rpm for 5 min (Eppendorf 5804 R). Te cell pellets were subjected to DNA extraction. A TAKARA DNA extraction kit was employed for DNA extraction. DNA quality was evaluated on a BASIC biospectrometer (Eppendorf).

DNA sequencing and assembly. Te genome of G. xylinus CGMCC 2955 was sequenced at Genewiz Co., Ltd. (Suzhou, China) by means of a PacBio RS II DNA sequencing 9 K library. Single-molecule real-time sequencing (SMRT) allows for highly precise sequencing, with accuracy of more than 99.999% (QV50) and is not afected by GC and AT content. Sequences were assembled according to principles similar to those of the frst-generation sequencing technology. Consequently, a single scafold was assembled in the SMRT Link sofware (version 4.1). Te genomic fne drawing was completed by analysing bioinformatic means afer quality control procedures.

Efects of acetate and ethanol on BC production. G. xylinus CGMCC 2955 was cultured as previously reported3. When acetate or ethanol was supplied as the sole carbon source, the culture media had the following composition: 0.43 (mol C)/l carbon source, 10 g/l peptone, 7.5 g/l yeast extract, and 10 g/l Na2HPO4. Te initial pH was adjusted to 6.0. Te same culture medium without addition of a carbon source served as the control. For supplementary ethanol or acetate experiments, 0.43 (mol C)/l ethanol or acetate was added to the glucose culture medium. BC was harvested afer static culture at 30 °C for 10 days. Before weighing, BC was processed as reported elsewhere3. Te dry weight was recorded for each pellicle at room temperature.

Nucleotide sequence accession numbers. Te complete sequences of G. xylinus CGMCC 2955 analysed in this study can be found in the NCBI GenBank (http://www.ncbi.nlm.nih.gov) under accession No. CP024644.

Biological detection of signalling molecule AHLs. Te presence of AHLs in G. xylinus CGMCC 2955 was determined in an agar well-difusion assay and β-galactosidase activity assay based on a biosensor46. Tis is a

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 7 www.nature.com/scientificreports/

Figure 5. Te metabolic and regulatory network of G. xylinus CGMCC 2955. Framed means regulation by ethanol and acetate addition, and boldfacing denotes regulation by the QS system. Gray means needing further examination. A solid line indicates that gene expression, metabolite synthesis, or metabolic pathway was activated; a dashed line means they were inhibited. (Abbreviations of genes. 6pgad: 6-phosphogluconate dehydrogenase; 6pgd: glucose-6-phosphate dehydrogenase; 6pgl: 6-phosphogluconolactonase; ack: acetate kinase; adh: alcohol dehydrogenase; aldh: aldehyde dehydrogenase; bcs: bacterial cellulose synthase; dgc: diguanylate cyclase; frk: ; g3pd: glycerol-3-phosphate dehydrogenase; gd: glucose dehydrogenase; gk: glucokinase; glyk: glycerol kinase; gntK: gluconokinase; ldh: lactate dehydrogenase; pats: phosphate acetyltransferase; pde: phosphodiesterase; pfB: phosphofructokinase B; pgm: phosphoglucomutase; pk: ; rpe: ribulose-phosphate 3-epimerase; tkt: transketolase; tpi: triosephosphate isomerase; tsd: transaldolase; UGPase: UDP-glucose pyrophosphorylase), (Abbreviations of metabolites. AC: acetate; ACCOA: acetyl-coenzyme A; ACD: acetaldehyde; BC: bacterial cellulose; c-di-GMP: cyclic diguanylate; DHAP: dihydroxyacetone phosphate; E4P: erythrose 4-phosphate; ETH: ethanol; F16P: fructose-1,6-diphosphate; F6P: fructose 6-phosphate; FRU: fructose; G1P: glucose 1-phosphate; G3P: glyceraldehyde-3-phosphate; G6P: glucose 6-phosphate; GAP: glyceraldehyde-3-phosphate; Glc: glucose; GlcA: gluconate; GlcA6P: 6-phosphogluconate; GlcL6P: 6-phosphogluconolactone; GLY: glycerol; GLY3P: glycerol 3-phosphate; GMP: guanosine monophosphate; GTP: guanosine triphosphate; LAT: lactate; OAA: oxaloacetate; PEP: phosphoenol pyruvate; PYR: pyruvate; RIB5P: ribose-5-phosphate; RIBU5P: ribulose-5-phosphate; SED7P: sedoheptulose 7-phosphate; UDPG: uridine-5′-phosphate-α-D-glucose; XYL5P: xylulose-5-phosphate).

traditional and powerful method for investigating QS systems in gram-negative bacteria. Te AHLs were detected by a vertical streaking agar well difusion assay with the AHL biosensor. G. xylinus CGMCC 2955 was streaked on Luria Bertani (LB) agar containing 40 μg/ml 5-bromo-4-chloro- 3-indolyl β-D-galactopyranoside (X-Gal) perpendicular to A. tumefaciens A136, with P. aeruginosa PAK serv- ing as the positive control and A. tumefaciens A136 alone serving as the negative control. AHL production was detected as the production of a blue pigment by A. tumefaciens A136. Te AHL biosensor, carrying TraR-regulated traI-lacZ fusion genes, can produce a blue pigment in the presence of X-Gal in response to exogenous AHLs.

Data availability statement. All data generated or analysed during this study are included in this pub- lished article.

Ethical statement. Tis article does not contain any experiments on human participants or animals that were performed by any of the authors. References 1. Koyama, M., Helbert, W., Imai, T., Sugiyama, J. & Henrissat, B. Parallel-up structure evidences the molecular directionality during biosynthesis of bacterial cellulose. Proc. Natl. Acad. Sci. USA 94, 9091–9095 (1997). 2. Ross, P., Mayer, R. & Benziman, M. Cellulose biosynthesis and function in bacteria. Microbiol. Rev. 55, 35–58 (1991). 3. Zhong, C. et al. Metabolic fux analysis of Gluconacetobacter xylinus for bacterial cellulose production. Appl. Microbiol. Biotechnol. 97, 6189–6199 (2013). 4. Hong, F. & Qiu, K. An alternative carbon source from konjac powder for enhancing production of bacterial cellulose in static cultures by a model strain Acetobacter aceti subsp. xylinus ATCC 23770. Carbohydr. Polym. 72, 545–549 (2008).

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 8 www.nature.com/scientificreports/

5. Lee, K. Y., Buldum, G., Mantalaris, A. & Bismarck, A. More than meets the eye in bacterial cellulose: biosynthesis, bioprocessing, and applications in advanced fber composites. Macromol. Biosci. 14, 10–32 (2014). 6. Hu, W., Chen, S., Yang, J., Li, Z. & Wang, H. Functionalized bacterial cellulose derivatives and nanocomposites. Carbohydr. Polym. 101, 1043–1060 (2014). 7. Cameron, D. E., Bashor, C. J. & Collins, J. J. A brief history of synthetic biology. Nat. Rev. Microbiol. 12, 381–390 (2014). 8. Florea, M. et al. Engineering control of bacterial cellulose production using a genetic toolkit and a new cellulose-producing strain. Proc. Natl. Acad. Sci. USA 113, E3431–E3440 (2016). 9. Kubiak, K. et al. Complete genome sequence of Gluconacetobacter xylinus E25 strain–valuable and efective producer of bacterial nanocellulose. J. Biotechnol. 176, 18–19 (2014). 10. Ogino, H. et al. Complete genome sequence of NBRC 3288, a unique cellulose-nonproducing strain of Gluconacetobacter xylinus isolated from vinegar. J. Biotechnol. 193, 6997–6998 (2011). 11. Zhang, H. et al. Complete genome sequence of the cellulose-producing strain Komagataeibacter nataicola RZS01. Sci. Rep. 7, 4431 (2017). 12. Jung, H. I. et al. Infuence of glycerol on production and structural-physical properties of cellulose from Acetobacter sp V6 cultured in shake fasks. Bioresour. Technol. 101, 3602–3608 (2010). 13. Shah, J. & Brown, R. M. Jr. Towards displays made from microbial cellulose. Appl. Microbiol. Biotechnol. 66, 352–355 (2005). 14. Fabra, M. J., López-Rubio, A., Ambrosio-Martín, J. & Lagaron, J. M. Improving the barrier properties of thermoplastic corn - based flms containing bacterial cellulose nanowhiskers by means of PHA electrospun coatings of interest in food packaging. Food Hydrocoll. 61, 261–268 (2016). 15. Czaja, W., Krystynowicz, A., Bielecki, S. & B, R. Jr. Microbial cellulose–the natural power to heal wounds. Biomaterials 27, 145–151 (2006). 16. Yin, N. et al. Bacterial cellulose as a substrate for microbial cell culture. Appl. Environ. Microbiol. 80, 1926–1932 (2014). 17. Iguchi, M., Yamanaka, S. & Budhiono, A. Bacterial cellulose-a masterpiece of nature’s arts. J. Mater. Sci. 35, 261–270 (2000). 18. Liu, M. et al. Metabolic investigation in Gluconacetobacter xylinus and its bacterial cellulose production under a direct current electric feld. Front. Microbiol. 7, 331 (2016). 19. Yang, X. N. et al. Improvement of antimicrobial activity of graphene oxide/bacterial cellulose nanocomposites through the electrostatic modifcation. Carbohydr. Polym. 136, 1152–1160 (2016). 20. Zhu, H. et al. Characterization of bacteriostatic sausage casing: A composite of bacterial cellulose embedded with ɛ-polylysine. Food Sci. Biotechnol. 19, 1479–1484 (2010). 21. Prust, C. et al. Complete genome sequence of the bacterium Gluconobacter oxydans. Nat. Biotechnol. 23, 195–200 (2005). 22. Illeghems, K., De, V. L. & Weckx, S. Complete genome sequence and comparative analysis of Acetobacter pasteurianus 386B, a strain well-adapted to the cocoa bean fermentation ecosystem. BMC Genomics 14, 526 (2013). 23. Liu, M. et al. Metabolomic profling coupled with metabolic network reveals diferences in Gluconacetobacter xylinus from static and agitated cultures. Biochem. Eng. J. 101, 85–98 (2015). 24. Naritomi, T., Kouda, T., Yano, H. & Yoshinaga, F. Efect of ethanol on bacterial cellulose production from fructose in continuous culture. J. Ferment. Bioeng. 85, 598–603 (1998). 25. Wanner, B. L. & Wilmesriesenberg, M. R. Involvement of phosphotransacetylase, acetate kinase, and acetyl phosphate synthesis in control of the phosphate regulon in Escherichia coli. J. Bacteriol. 174, 2124–2130 (1992). 26. Ingledew, W. J. & Poole, R. K. Te respiratory chains of Escherichia coli. Microbiol. Rev. 48, 222–271 (1984). 27. Ji, K. et al. Bacterial cellulose synthesis mechanism of facultative anaerobe Enterobacter sp. FY-07. Sci. Rep. 6, 21863 (2016). 28. Imai, T., Sun, S., Horikawa, Y., Wada, M. & Sugiyama, J. Functional reconstitution of cellulose synthase in Escherichia coli. Biomacromolecules 15, 4206–4213 (2014). 29. Kawano, S. et al. Cloning of cellulose synthesis related genes from Acetobacter xylinum ATCC23769 and ATCC53582: comparison of cellulose synthetic ability between strains. DNA Res. 9, 149–156 (2002). 30. Hu, S. Q. & Cosgrove, D. J. Structure of bacterial cellulose synthase subunit D octamer with four inner passageways. Proc. Natl. Acad. Sci. USA 107, 17957–17961 (2010). 31. Omadjela, O. et al. BcsA and BcsB form the catalytically active core of bacterial cellulose synthase sufcient for in vitro cellulose synthesis. Proc. Natl. Acad. Sci. USA 110, 17856–17861 (2013). 32. Wong, H. C. et al. Genetic organization of the cellulose synthase operon in Acetobacter xylinum. Proc. Natl. Acad. Sci. USA 87, 8130–8134 (1990). 33. Römling, U. & Galperin, M. Y. Bacterial cellulose biosynthesis: diversity of operons, subunits, products, and functions. Trends Microbiol. 23, 545–557 (2015). 34. Koo, H. M., Song, S. H., Yu, R. P. & Yu, S. K. Evidence that a β-1,4-endoglucanase secreted by Acetobacter xylinum plays an essential role for the formation of cellulose fber. Biosci. Biotechnol. Biochem. 62, 2257–2259 (1998). 35. Nakai, T. et al. Formation of highly twisted ribbons in a carboxymethylcellulase gene-disrupted strain of a cellulose-producing bacterium. J. Bacteriol. 195, 958–964 (2013). 36. Umeda, Y. et al. Cloning of cellulose synthase genes from Acetobacter xylinum JCM 7664: implication of a novel set of cellulose synthase genes. DNA Res. 6, 109–115 (1999). 37. Morgan, J. L. W., Mcnamara, J. T. & Zimmer, J. Mechanism of activation of bacterial cellulose synthase by cyclic-di-GMP. Nat. Struct. Mol. Biol. 21, 489–498 (2014). 38. Ross, P. et al. Control of cellulose synthesis Acetobacter xylinum. A unique guanyl oligonucleotide is the immediate activator of the cellulose synthase. Carbohydr. Res. 149, 101–117 (1986). 39. Iida, A., Ohnishi, Y. & Horinouchi, S. Control of acetic acid fermentation by quorum sensing via N-acylhomoserine lactones in Gluconacetobacter intermedius. J. Bacteriol. 190, 2546–2555 (2008). 40. Iida, A., Ohnishi, Y. & Horinouchi, S. Identifcation and characterization of target genes of the GinI/GinR quorum-sensing system in Gluconacetobacter intermedius. Microbiology 155, 3021–3032 (2009). 41. Waters, C. M. & Bassler, B. L. Quorum sensing: cell-to-cell communication in bacteria. Annu. Rev. Cell Dev. Bio. 21, 319–346 (2005). 42. Fuqua, C. & Greenberg, E. P. Self perception in bacteria: quorum sensing with acylated homoserine lactones. Curr. Opin. Microbiol. 1, 183–189 (1998). 43. Ryan, R. P., An, S. Q., Allan, J. H., Mccarthy, Y. & Dow, J. M. Te DSF family of cell–cell signals: an expanding class of bacterial virulence regulators. Plos Pathog. 11, e1004986 (2015). 44. Tiaden, A., Spirig, T. & Hilbi, H. Bacterial gene regulation by alpha-hydroxyketone signaling. Trends Microbiol. 18, 288–297 (2010). 45. Shaw, P. D. et al. Detecting and characterizing N-acyl-homoserine lactone signal molecules by thin-layer chromatography. Proc. Natl. Acad. Sci. USA 94, 6036–6041 (1997). 46. Yeon, K. M. et al. Quorum sensing: a new biofouling control paradigm in a membrane for advanced wastewater treatment. Environ. Sci. Technol. 43, 380–385 (2009). 47. Schaefer, A. L. et al. E. P. Generation of cell-to-cell signals in quorum sensing: acyl homoserine lactone synthase activity of a purifed Vibrio fscheri LuxI protein. Proc. Natl. Acad. Sci. USA 93, 9505–9509 (1996). 48. Li, Y. et al. Improvement of bacterial cellulose production by manipulating the metabolic pathways in which ethanol and sodium citrate involved. Appl. Microbiol. Biotechnol. 96, 1479–1487 (2012).

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 9 www.nature.com/scientificreports/

49. Mikkelsen, D., Flanagan, B. M., Dykes, G. A. & Gidley, M. J. Infuence of diferent carbon sources on bacterial cellulose production by Gluconacetobacter xylinus strain ATCC 53524. J. Appl. Microbiol. 107, 576–583 (2010). 50. Schramm, M., Gromet, Z. & Hestrin, S. Role of phosphate in synthesis of cellulose by Acetobacter xylinum. Nature 179, 28–29 (1957). 51. Daldal, F. Molecular cloning of the gene for phosphofructokinase-2 of Escherichia coli and the nature of a mutation, pfkB1, causing a high level of the enzyme. J. Mol. Bio. 168, 285–305 (1983). 52. Blangy, D., Buc, H. & Monod, J. Kinetics of the allosteric interactions of phosphofructokinase from Escherichia coli. J. Mol. Bio. 31, 13–35 (1968). 53. Liu, M. et al. Enhanced bacterial cellulose production by Gluconacetobacter xylinus via expression of Vitreoscilla hemoglobin and oxygen tension regulation. Appl. Microbiol. Biotechnol. 102, 1155–1165 (2017). 54. Iyer, P. R., Geib, S. M., Catchmark, J., Kao, T. & Ming, T. Genome sequence of a cellulose-producing bacterium, Gluconacetobacter hansenii ATCC 23769. J. Bacteriol. 192, 4256 (2010). 55. Florea, M., Reeve, B., Abbott, J., Freemont, P. S. & Ellis, T. Genome sequence and plasmid transformation of the model high-yield bacterial cellulose producer Gluconacetobacter hansenii ATCC 53582. Sci. Rep. 6, 23635 (2016). 56. Nealson, K. H. & Hastings, J. W. Bacterial bioluminescence: its control and ecological signifcance. Microbiol. Rev. 43, 496–518 (1979). 57. Miller, M. B., Skorupski, K., Lenz, D. H., Taylor, R. K. & Bassler, B. L. Parallel quorum sensing systems converge to regulate virulence in . Cell 110, 303–314 (2002). 58. Claudia, A., Ursula, S. & Kirsten, J. Single cell analysis of Vibrio harveyi uncovers functional heterogeneity in response to quorum sensing signals. BMC Microbiol. 12, 1–10 (2012). 59. Ng, W. L. & Bassler, B. L. Bacterial quorum-sensing network architectures. Ann. Rev. Genet. 43, 197–222 (2009). 60. Lee, R. et al. A quorum-sensing antagonist targets both membrane-bound and cytoplasmic receptors and controls bacterial pathogenicity. Mol. Cell 35, 143–153 (2009). 61. Hawver, L. A., Jung, S. A. & Ng, W.-L. Specifcity and complexity in bacterial quorum-sensing systems. FEMS Microbiol. Rev. 40, 738–752 (2016). 62. Eberhard, A. et al. Structural identifcation of autoinducer of Photobacterium fscheri luciferase. Biochemistry 20, 2444–2449 (1981). 63. Engebrecht, J. & Silverman, M. Identifcation of genes and gene products necessary for bacterial bioluminescence. Proc. Natl. Acad. Sci. USA 81, 4154–4158 (1984). 64. Williams, W. S. & Cannon, R. E. Alternative environmental roles for cellulose produced by Acetobacter xylinum. Appl. Environ. Microbiol. 55, 2448–2452 (1989). 65. Parsek, M. R. & Greenberg, E. P. Sociomicrobiology: the connections between quorum sensing and bioflms. Trends Microbiol. 13, 27–33 (2005). Acknowledgements Te authors would like to express their appreciation to professor Robert J.C. McLean at Texas A&M University for supplying A. tumefaciens A136 and for fnancial support from the National Natural Science Foundation of China (grants No. 21576212 and No. 31470610), Natural Science Foundation of Tianjin (grant No. 15JCZDJC32600), and Innovation Foundation for Doctoral Dissertation of Tianjin University of Science and Technology (grant No. 2016001). Author Contributions Miao Liu and Lingpu Liu contributed to the annotation of the genome and analysed the genomic information; Miao Liu wrote the frst draf of the manuscript; Shiru Jia and Cheng Zhong contributed genomic analysis and manuscript writing suggestions; Siqi Li and Yang Zou substantially participated in most experiments as well as manuscript revisions. Additional Information Competing Interests: Te authors declare no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. Te images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

© Te Author(s) 2018

ScIEnTIFIc ReportS | (2018)8:6266 | DOI:10.1038/s41598-018-24559-w 10